A NEW APPROACH TO MODELING CONDITION-BASED MAINTENANCE FOR STOCHASTICALLY DETERIORATING SYSTEMS

Abstract

This paper develops the degradation and maintenance modeling technique for repairable systems subject to deterioration due to aging and damage caused by operating environment factors. This provides a relaxed and generalized approach to modeling condition-based maintenance. The approach can deal with both the maintenance scheduling problem and failures preventing the system functioning further. Controlled by a virtual age process and a general class of process called Piecewise-Stochastic (Damage) process (PSP), the proportional intensity model (PIM) is used to describe the system state. The state of the system is monitored at periodic times and maintenance actions are carried out in response to the system state revealed at inspection times. Assuming a threshold-type policy, the approach aims at minimizing the long-run average cost per unit time subject to maintenance parameters; the inspection interval and preventive replacement threshold. To this end, given some assumptions, expressions for the expected cycle length and expected cost per cycle emerging as solutions of the Fredholm integral equations are obtained. The solution technique has been presented for the case when the effect of operating environment is modeled as a Gamma process.